Novel time domain multi-class SVMs for landmark detection

Keyphrases

• Landmark Detection 100%
• One-class Support Vector Machine (OCSVM) 100%
• Phoneme Boundary 66%
• Acoustics 33%
• Alignment Error 33%
• Speech Recognition 33%
• Accurate Segmentation 33%
• Benchmark Dataset 33%
• Acoustic Features 33%
• Phoneme 33%
• Spoken Language 33%
• Speech Recognizer 33%
• Art Performance 33%
• Speech Recognition Model 33%
• Phone Boundary 33%
• Time Alignment 33%
• HMM-based 33%
• Boundary Alignment 33%
• Training Corpus 33%
• SVM-based 33%
• Telugu 33%
• Indian Languages 33%
• Binary SVM 33%

Computer Science

• Landmark Detection 100%
• Support Vector Machine 100%
• Speech Recognition 66%
• Acoustic Feature 33%
• New-State 33%
• Art Performance 33%
• Speech Recognizer 33%
• Spoken Language 33%